CN108693441A - A kind of electric transmission line isolator recognition methods and system - Google Patents
A kind of electric transmission line isolator recognition methods and system Download PDFInfo
- Publication number
- CN108693441A CN108693441A CN201810338067.4A CN201810338067A CN108693441A CN 108693441 A CN108693441 A CN 108693441A CN 201810338067 A CN201810338067 A CN 201810338067A CN 108693441 A CN108693441 A CN 108693441A
- Authority
- CN
- China
- Prior art keywords
- transmission line
- model
- image
- images
- aerial images
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
- G01R31/08—Locating faults in cables, transmission lines, or networks
- G01R31/081—Locating faults in cables, transmission lines, or networks according to type of conductors
- G01R31/085—Locating faults in cables, transmission lines, or networks according to type of conductors in power transmission or distribution lines, e.g. overhead
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
- G01R31/08—Locating faults in cables, transmission lines, or networks
- G01R31/088—Aspects of digital computing
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Theoretical Computer Science (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a kind of electric transmission line isolator recognition methods and systems.This method includes:Obtain analog image and historical background image, the analog image is the simulation Aerial Images of transmission line of electricity, the analog image for no background insulation subgraph;Neural network model is trained by the analog image and the historical background image, obtains Classification and Identification model;Transmission line of electricity is shot, multiple current Aerial Images of transmission line of electricity are obtained;Color division is carried out to the multiple current Aerial Images by hexagonal vertebral model, obtains the pre-selected images of different colour systems;According to the electric transmission line isolator in the pre-selected images of different colour systems described in the Classification and Identification Model Identification.The method or system provided through the invention can accurately and rapidly identify the insulator of transmission line of electricity.
Description
Technical field
The present invention relates to insulators to identify field, more particularly to a kind of electric transmission line isolator recognition methods and system.
Background technology
For insulator as the key equipment in transmission line of electricity, status monitoring is particularly important to the stable operation of power grid.Absolutely
Edge is one of the important equipment in overhead transmission line rack, and playing mechanical support and electric insulation prevents the work on electric current time ground
With, while being also one of Frequent Troubles element, the failures such as damaged, self-destruction, crackle and foreign matter seriously threaten the peace of transmission line of electricity
Full reliability service.According to statistics, in the transmission line of electricity accident occurred at present, trip event caused by insulator breakdown accounts for 81.3%.
Therefore, the failure of insulator is particularly significant in investigation transmission line of electricity in time.
Invention content
The object of the present invention is to provide a kind of electric transmission line isolator recognition methods and systems, accurately and rapidly to know
The insulator of other transmission line of electricity, the failure for insulator in investigation transmission line of electricity road in time provide safeguard.
To achieve the above object, the present invention provides following schemes:
A kind of electric transmission line isolator recognition methods, the method includes:
Analog image and historical background image are obtained, the analog image is the simulation Aerial Images of transmission line of electricity, described
Analog image for no background insulation subgraph;
Neural network model is trained by the analog image and the historical background image, obtains Classification and Identification
Model;
Transmission line of electricity is shot, multiple current Aerial Images of transmission line of electricity are obtained;
Color division is carried out to the multiple current Aerial Images by hexagonal vertebral model, obtains the pre-selection of different colour systems
Image;
According to the electric transmission line isolator in the pre-selected images of different colour systems described in the Classification and Identification Model Identification.
Optionally, described that neural network model is trained by the analog image, Classification and Identification model is obtained, is had
Body includes:
Classified to the analog image and the historical background image by the neural network model, obtains background
Figure output data and insulation subgraph output data;
Judge the Background output data and insulation sub-image data whether within the scope of error threshold;
It is no to be, it is determined that the neural network model is Classification and Identification model;
If it is not, then adjusting the parameter of the neural network model, make the Background output data and insulation subgraph
Data obtain Classification and Identification model within the scope of error threshold.
Optionally, described that color division is carried out to the multiple current Aerial Images by hexagonal vertebral model, it obtains not
With the pre-selected images of colour system, specifically include:
The value range of tone, saturation degree and lightness in acquisition hexagonal vertebral model;
According to the value range of tone, saturation degree and lightness in the hexagonal vertebral model, to the multiple current
Aerial Images carry out color division, obtain the pre-selected images of different colour systems.
Optionally, the tone according in the hexagonal vertebral model, saturation degree and lightness value range, to institute
It states multiple current Aerial Images and carries out color division, obtain the pre-selected images of different colour systems, specifically include:
Obtain the pixel value of each pixel in each current Aerial Images;
According in the pixel value of each pixel in each current Aerial Images and the hexagonal vertebral model tone,
The value range of saturation degree and lightness determines the corresponding colour system of each current Aerial Images, obtains the pre-selected images of different colour systems.
A kind of electric transmission line isolator identifying system, the system comprises:
Image collection module, for obtaining analog image and historical background image, the analog image is transmission line of electricity
Simulate Aerial Images, the analog image for no background insulation subgraph;
Training module, for being instructed to neural network model by the analog image and the historical background image
Practice, obtains Classification and Identification model;
Taking module obtains multiple current Aerial Images of transmission line of electricity for being shot to transmission line of electricity;
Pre-selected images division module is drawn for carrying out color to the multiple current Aerial Images by hexagonal vertebral model
Point, obtain the pre-selected images of different colour systems;
Identification module, it is exhausted for the transmission line of electricity in the pre-selected images of different colour systems described in the Classification and Identification Model Identification
Edge.
Optionally, the training module includes:
Taxon, for being carried out to the analog image and the historical background image by the neural network model
Classification obtains Background output data and insulation subgraph output data;
Judging unit, for judging the Background output data and insulation sub-image data whether in error threshold model
In enclosing;
Determination unit is connect with the judgment module, for working as the Background output data and insulator picture number
When according within the scope of error threshold, determine that the neural network model is Classification and Identification model;And for working as the Background
When output data and insulation sub-image data be not within the scope of error threshold, the parameter of the neural network model is adjusted, is made
The Background output data and insulation sub-image data obtain Classification and Identification model within the scope of error threshold.
Optionally, the pre-selected images division module includes:
Value range acquiring unit, for obtaining the tone in hexagonal vertebral model, saturation degree and the value model of lightness
It encloses;
Pre-selected images division unit, for being taken according to the tone in the hexagonal vertebral model, saturation degree and lightness
It is worth range, color division is carried out to the multiple current Aerial Images, obtains the pre-selected images of different colour systems.
Optionally, the pre-selected images division unit includes:
Pixel value obtains subelement, the pixel value for obtaining each pixel in each current Aerial Images;
Subelement is divided, the value range according to tone, saturation degree and lightness in the hexagonal vertebral model is used for,
Color division is carried out to the multiple current Aerial Images, obtains the pre-selected images of different colour systems.
According to specific embodiment provided by the invention, the invention discloses following technique effects:The present invention is for figure of taking photo by plane
The identification location technology of insulator as in is studied, and carries out target classification identification using neural network model, and pass through face
Color model obtains pre-selected images, it is ensured that the accuracy of preselected area segmentation orientation range before carrying out Classification and Identification, to carry
The recognition accuracy of high-class identification model accurately and rapidly identifies the insulator of transmission line of electricity, for investigation transmission line of electricity in time
The failure of insulator provides safeguard in road.
Description of the drawings
It in order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, below will be to institute in embodiment
Attached drawing to be used is needed to be briefly described, it should be apparent that, the accompanying drawings in the following description is only some implementations of the present invention
Example, for those of ordinary skill in the art, without having to pay creative labor, can also be according to these attached drawings
Obtain other attached drawings.
Fig. 1 is the flow chart of electric transmission line isolator recognition methods of the embodiment of the present invention;
Fig. 2 is the structure diagram of electric transmission line isolator identifying system of the embodiment of the present invention.
Specific implementation mode
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation describes, it is clear that described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.It is based on
Embodiment in the present invention, it is obtained by those of ordinary skill in the art without making creative efforts every other
Embodiment shall fall within the protection scope of the present invention.
The object of the present invention is to provide a kind of electric transmission line isolator recognition methods and systems, accurately and rapidly to know
The insulator of other transmission line of electricity, the failure for insulator in investigation transmission line of electricity road in time provide safeguard.
In order to make the foregoing objectives, features and advantages of the present invention clearer and more comprehensible, below in conjunction with the accompanying drawings and specific real
Applying mode, the present invention is described in further detail.
Fig. 1 is the flow chart of electric transmission line isolator recognition methods of the embodiment of the present invention.As shown in Figure 1, a kind of power transmission line
Road insulator recognition methods mainly includes the following steps that:
Step 101:Obtain analog image and historical background image, the analog image is that the simulation of transmission line of electricity is taken photo by plane figure
Picture, the analog image for no background insulation subgraph.
Specifically, according to the size and shape of insulator, on preceding, top, left etc., different views carries out insulation interest respectively
Belong to end, umbrella disk and stick core drafting, due to its drafting be threedimensional model, it is therefore desirable in different views to its into
Row adjustment.Respectively all parts of insulator are carried out with the setting of material parameters.All parts are combined alignment.Carry out wash with watercolours
The setting of environment is contaminated, finished product is rendered.It is arranged according to angle, generates comprehensive analog image.
Step 102:Neural network model is trained by the analog image and the historical background image, is obtained
Classification and Identification model.
Specifically, classifying to the analog image by the neural network model, Background output data is obtained
And insulation subgraph output data;
Judge the Background output data and insulation sub-image data whether within the scope of error threshold;
It is no to be, it is determined that the neural network model is Classification and Identification model;
If it is not, then adjusting the parameter of the neural network model, make the Background output data and insulation subgraph
Data obtain Classification and Identification model within the scope of error threshold.
We open insulator simulation artificial image 20000 as the positive sample in sample database, do not include in Aerial Images
The real background image 20000 of insulator is opened as the negative sample in sample database.Sample size is 256 × 256, in sample database
Sample distribution is 1:1.
Neural network model is made of 3 groups of convolutional layers and down-sampling layer and 1 full articulamentum.Convolutional layer conv1
By using 256 7 × 7 convolution kernel sliding processing 256 × 256 × 3 by altimetric image;The receptive field of down-sampling layer pool2
Size is 2 × 2;Convolutional layer conv3 and conv5 use 128 6 × 6 convolution kernels and 64 3 × 3 convolution kernels respectively;Under adopt
The receptive field size of sample layer pool4 and pool6 are all 3 × 3;Whole network training parameter number shares 65280.
We use mean square deviation as our loss function in the method, as shown in Equation 2.
In formula:L is loss function value;N is batch processing quantity, and present networks take 150;W is network weight, and λ r (W) are canonical
;f(zit) it is that sample exports zitIt is general to be classified the corresponding classification of sample for corresponding t class probabilities in final output
Rate.zijFor the corresponding output valve of jth class, zimaxFor input sample XiMaximum value in output.
Network weight more new formula is as follows:
Wt+1=Wt+μVt-α▽L(Wt) (4)
α × 0.1 α=base_floor(iter/5000)(5)
Wherein Wt+1For the weights of t+1 wheels;WtFor the weights of t wheels;μ is the weight of last Grad;VtIt is taken turns for t
The updated value of weight;Derivation for loss function to weights;α is learning rate, is updated according to formula (5),
Base_ α are learning rate initial values 0.001, and iter is current iterations.
Step 103:Transmission line of electricity is shot, multiple current Aerial Images of transmission line of electricity are obtained.
Step 104:Color division is carried out to the multiple current Aerial Images by hexagonal vertebral model, is obtained not homochromy
The pre-selected images of system.
Color is the most obvious characteristic mutually distinguished between object, has important work in the image understanding based on content
With.Currently, the existing method for indicating color of image feature includes mainly:Color histogram, color moment and color correlogram etc..
Color histogram mainly counts the color pixel values in image using statistical method;Color moment is by all pictures
The color value of vegetarian refreshments regards a probability distribution as, then this discrete features is indicated by its each rank square.Color correlogram
It is then that the spatial relationship between color histogram and color is combined carry out character representation.These methods are only counting
The color characteristic of image is demonstrated by, there is no really from the enterprising row information expression of semantic content.The most direct content of color
Information is the title of color.In natural language, there are the noun of many description colors, different language poor to the classification of color
It is very not big.This method will likely appear in the color in the Aerial Images of transmission line of electricity road and be divided into eight colour systems, respectively grey
System, white color system, red colour system, yellow class, green system, cyan system, blue series and violet.
Color division is carried out to the multiple current Aerial Images by hexagonal vertebral model and the colour system group, is obtained
The pre-selected images of different colour systems, specifically include:
The value range of tone H, saturation degree S and lightness V in acquisition hexagonal vertebral model;
According in the pixel value of each pixel in each current Aerial Images and the hexagonal vertebral model tone,
The value range of saturation degree and lightness determines the corresponding colour system of each current Aerial Images, obtains the pre-selected images of different colour systems.
Table 1 is the value range that each cie system of color representation cie corresponds to H, S, V variable.
1 each cie system of color representation cie of table corresponds to the value range of H, S, V variable
The pixel for meeting above-mentioned prescribed limit is classified as corresponding cie system of color representation cie, in preselected area division, each face
Colour system corresponds to width pre-selection figure, and in preselecting figure, the pixel value of only corresponding color system pixel is remained stationary, and rest of pixels value is set
The Xiang Suzhi [ of background picture when being generated for training sample;224,224,224].
Step 105:It is exhausted according to the transmission line of electricity in the pre-selected images of different colour systems described in the Classification and Identification Model Identification
Edge.
According to specific embodiment provided by the invention, the invention discloses following technique effects:The present invention is for figure of taking photo by plane
The identification location technology of insulator as in is studied, and carries out target classification identification using neural network model, and pass through face
Color model obtains pre-selected images, it is ensured that the accuracy of preselected area segmentation orientation range before carrying out Classification and Identification, to carry
The recognition accuracy of high-class identification model accurately and rapidly identifies the insulator of transmission line of electricity, for investigation transmission line of electricity in time
The failure of insulator provides safeguard in road.
Fig. 2 is the structure diagram of electric transmission line isolator identifying system of the embodiment of the present invention.As shown in Fig. 2, a kind of transmission of electricity
Line insulator identifying system includes:
Image collection module 201, for obtaining analog image and historical background image, the analog image is transmission line of electricity
Simulation Aerial Images, the analog image for no background insulation subgraph.
Training module 202, for being carried out to neural network model by the analog image and the historical background image
Training, obtains Classification and Identification model.
The training module 202 specifically includes:
Taxon, for being carried out to the analog image and the historical background image by the neural network model
Classification obtains Background output data and insulation subgraph output data;
Judging unit, for judging the Background output data and insulation sub-image data whether in error threshold model
In enclosing;
Determination unit is connect with the judgment module, for working as the Background output data and insulator picture number
When according within the scope of error threshold, determine that the neural network model is Classification and Identification model;And for working as the Background
When output data and insulation sub-image data be not within the scope of error threshold, the parameter of the neural network model is adjusted, is made
The Background output data and insulation sub-image data obtain Classification and Identification model within the scope of error threshold.
Taking module 203 obtains multiple current Aerial Images of transmission line of electricity for being shot to transmission line of electricity.
Pre-selected images division module 204, for carrying out face to the multiple current Aerial Images by hexagonal vertebral model
Color divides, and obtains the pre-selected images of different colour systems.
The pre-selected images division module 204 specifically includes:
Value range acquiring unit, for obtaining the tone in hexagonal vertebral model, saturation degree and the value model of lightness
It encloses;
Pre-selected images division unit, for being taken according to the tone in the hexagonal vertebral model, saturation degree and lightness
It is worth range, color division is carried out to the multiple current Aerial Images, obtains the pre-selected images of different colour systems.The pre-selected images
Division unit includes:
Pixel value obtains subelement, the pixel value for obtaining each pixel in each current Aerial Images;
Subelement is divided, the pixel value for each pixel in each current Aerial Images of basis and the hexagonal vertebra
The value range of tone, saturation degree and lightness in body Model determines the corresponding colour system of each current Aerial Images, obtains difference
The pre-selected images of colour system.
Identification module 205, for the power transmission line in the pre-selected images of different colour systems described in the Classification and Identification Model Identification
Road insulator.
Each embodiment is described by the way of progressive in this specification, the highlights of each of the examples are with other
The difference of embodiment, just to refer each other for identical similar portion between each embodiment.For system disclosed in embodiment
For, since it is corresponded to the methods disclosed in the examples, so description is fairly simple, related place is said referring to method part
It is bright.
Principle and implementation of the present invention are described for specific case used herein, and above example is said
The bright method and its core concept for being merely used to help understand the present invention;Meanwhile for those of ordinary skill in the art, foundation
The thought of the present invention, there will be changes in the specific implementation manner and application range.In conclusion the content of the present specification is not
It is interpreted as limitation of the present invention.
Claims (8)
1. a kind of electric transmission line isolator recognition methods, which is characterized in that the method includes:
Analog image and historical background image are obtained, the analog image is the simulation Aerial Images of transmission line of electricity, the simulation
Image for no background insulation subgraph;
Neural network model is trained by the analog image and the historical background image, obtains Classification and Identification mould
Type;
Transmission line of electricity is shot, multiple current Aerial Images of transmission line of electricity are obtained;
Color division is carried out to the multiple current Aerial Images by hexagonal vertebral model, obtains the pre-selection figure of different colour systems
Picture;
According to the electric transmission line isolator in the pre-selected images of different colour systems described in the Classification and Identification Model Identification.
2. recognition methods according to claim 1, which is characterized in that it is described by the analog image to neural network mould
Type is trained, and is obtained Classification and Identification model, is specifically included:
Classified to the analog image and the historical background image by the neural network model, it is defeated to obtain Background
Go out data and insulation subgraph output data;
Judge the Background output data and insulation sub-image data whether within the scope of error threshold;
It is no to be, it is determined that the neural network model is Classification and Identification model;
If it is not, then adjusting the parameter of the neural network model, make the Background output data and insulation sub-image data
Within the scope of error threshold, Classification and Identification model is obtained.
3. recognition methods according to claim 1, which is characterized in that described to be worked as to the multiple by hexagonal vertebral model
Preceding Aerial Images carry out color division, obtain the pre-selected images of different colour systems, specifically include:
The value range of tone, saturation degree and lightness in acquisition hexagonal vertebral model;
According to the value range of tone, saturation degree and lightness in the hexagonal vertebral model, currently take photo by plane to the multiple
Image carries out color division, obtains the pre-selected images of different colour systems.
4. recognition methods according to claim 3, which is characterized in that the color according in the hexagonal vertebral model
It adjusts, the value range of saturation degree and lightness, color division is carried out to the multiple current Aerial Images, obtains different colour systems
Pre-selected images specifically include:
Obtain the pixel value of each pixel in each current Aerial Images;
According to the tone in the pixel value of each pixel in each current Aerial Images and the hexagonal vertebral model, saturation
The value range of degree and lightness determines the corresponding colour system of each current Aerial Images, obtains the pre-selected images of different colour systems.
5. a kind of electric transmission line isolator identifying system, which is characterized in that the system comprises:
Image collection module, for obtaining analog image and historical background image, the analog image is the simulation of transmission line of electricity
Aerial Images, the analog image for no background insulation subgraph;
Training module is obtained for being trained to neural network model by the analog image and the historical background image
To Classification and Identification model;
Taking module obtains multiple current Aerial Images of transmission line of electricity for being shot to transmission line of electricity;
Pre-selected images division module, for carrying out color division to the multiple current Aerial Images by hexagonal vertebral model,
Obtain the pre-selected images of different colour systems;
Identification module, for the transmission line insulator in the pre-selected images of different colour systems described in the Classification and Identification Model Identification
Son.
6. system according to claim 5, which is characterized in that the training module includes:
Taxon, for being divided the analog image and the historical background image by the neural network model
Class obtains Background output data and insulation subgraph output data;
Judging unit, for judging the Background output data and insulation sub-image data whether in error threshold range
It is interior;
Determination unit is connect with the judgment module, for existing when the Background output data and insulation sub-image data
When within the scope of error threshold, determine that the neural network model is Classification and Identification model;And for being exported when the Background
When data and insulation sub-image data be not within the scope of error threshold, the parameter of the neural network model is adjusted, is made described
Background output data and insulation sub-image data obtain Classification and Identification model within the scope of error threshold.
7. system according to claim 5, which is characterized in that the pre-selected images division module includes:
Value range acquiring unit, the value range for obtaining the tone in hexagonal vertebral model, saturation degree and lightness;
Pre-selected images division unit, for the value model according to tone, saturation degree and lightness in the hexagonal vertebral model
It encloses, color division is carried out to the multiple current Aerial Images, obtains the pre-selected images of different colour systems.
8. system according to claim 5, which is characterized in that the pre-selected images division unit includes:
Pixel value obtains subelement, the pixel value for obtaining each pixel in each current Aerial Images;
Subelement is divided, the pixel value for each pixel in each current Aerial Images of basis and the hexagonal centrum mould
The value range of tone, saturation degree and lightness in type determines the corresponding colour system of each current Aerial Images, obtains different colour systems
Pre-selected images.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810338067.4A CN108693441B (en) | 2018-04-16 | 2018-04-16 | Method and system for identifying insulator of power transmission line |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810338067.4A CN108693441B (en) | 2018-04-16 | 2018-04-16 | Method and system for identifying insulator of power transmission line |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108693441A true CN108693441A (en) | 2018-10-23 |
| CN108693441B CN108693441B (en) | 2021-01-22 |
Family
ID=63845651
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810338067.4A Active CN108693441B (en) | 2018-04-16 | 2018-04-16 | Method and system for identifying insulator of power transmission line |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108693441B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110531218A (en) * | 2019-09-04 | 2019-12-03 | 国网四川省电力公司乐山供电公司 | A kind of transmission line malfunction identifying system based on image outline detection |
| CN112270234A (en) * | 2020-10-20 | 2021-01-26 | 天津大学 | Power transmission line insulation sub-target identification method based on aerial image |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1271095A (en) * | 2000-03-14 | 2000-10-25 | 华中理工大学 | Automatic product image detecting system |
| CN101436302A (en) * | 2008-12-10 | 2009-05-20 | 南京大学 | Method for sorting colors of colorful three-dimensional model based on neural network |
| CN103544473A (en) * | 2013-09-17 | 2014-01-29 | 广东工业大学 | Electronic connector detection method based on machine vision |
| US9384560B2 (en) * | 2014-10-27 | 2016-07-05 | King Fahd University Of Petroleum And Minerals | Contamination level estimation method for high voltage insulators |
| CN105956524A (en) * | 2016-04-22 | 2016-09-21 | 北京智芯原动科技有限公司 | Method and device for identifying traffic signs |
| CN106919932A (en) * | 2017-03-13 | 2017-07-04 | 华北电力大学（保定） | A kind of insulator of " simulation is true " parallel construction positions soft recognition methods |
| CN106934418A (en) * | 2017-03-09 | 2017-07-07 | 国家电网公司 | A kind of insulator infrared diagnostics method based on convolution Recursive Networks |
| CN107392901A (en) * | 2017-07-24 | 2017-11-24 | 国网山东省电力公司信息通信公司 | A kind of method for transmission line part intelligence automatic identification |
| CN107680090A (en) * | 2017-10-11 | 2018-02-09 | 电子科技大学 | Based on the electric transmission line isolator state identification method for improving full convolutional neural networks |
-
2018
- 2018-04-16 CN CN201810338067.4A patent/CN108693441B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1271095A (en) * | 2000-03-14 | 2000-10-25 | 华中理工大学 | Automatic product image detecting system |
| CN101436302A (en) * | 2008-12-10 | 2009-05-20 | 南京大学 | Method for sorting colors of colorful three-dimensional model based on neural network |
| CN103544473A (en) * | 2013-09-17 | 2014-01-29 | 广东工业大学 | Electronic connector detection method based on machine vision |
| US9384560B2 (en) * | 2014-10-27 | 2016-07-05 | King Fahd University Of Petroleum And Minerals | Contamination level estimation method for high voltage insulators |
| CN105956524A (en) * | 2016-04-22 | 2016-09-21 | 北京智芯原动科技有限公司 | Method and device for identifying traffic signs |
| CN106934418A (en) * | 2017-03-09 | 2017-07-07 | 国家电网公司 | A kind of insulator infrared diagnostics method based on convolution Recursive Networks |
| CN106919932A (en) * | 2017-03-13 | 2017-07-04 | 华北电力大学（保定） | A kind of insulator of " simulation is true " parallel construction positions soft recognition methods |
| CN107392901A (en) * | 2017-07-24 | 2017-11-24 | 国网山东省电力公司信息通信公司 | A kind of method for transmission line part intelligence automatic identification |
| CN107680090A (en) * | 2017-10-11 | 2018-02-09 | 电子科技大学 | Based on the electric transmission line isolator state identification method for improving full convolutional neural networks |
Non-Patent Citations (2)
| Title |
|---|
| 程海燕: "一种电网巡检航拍图像中绝缘子定位方法", 《系统仿真学报》 * |
| 翟永杰: "基于骨架提取的航拍绝缘子图像分步识别方法", 《华北电力大学学报》 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110531218A (en) * | 2019-09-04 | 2019-12-03 | 国网四川省电力公司乐山供电公司 | A kind of transmission line malfunction identifying system based on image outline detection |
| CN110531218B (en) * | 2019-09-04 | 2021-07-27 | 国网四川省电力公司乐山供电公司 | Power transmission line fault recognition system based on image contour detection |
| CN112270234A (en) * | 2020-10-20 | 2021-01-26 | 天津大学 | Power transmission line insulation sub-target identification method based on aerial image |
| CN112270234B (en) * | 2020-10-20 | 2022-04-19 | 天津大学 | Power transmission line insulation sub-target identification method based on aerial image |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108693441B (en) | 2021-01-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107563422B (en) | A kind of polarization SAR classification method based on semi-supervised convolutional neural networks | |
| US11176418B2 (en) | Model test methods and apparatuses | |
| CN103839065B (en) | Extraction method for dynamic crowd gathering characteristics | |
| WO2021073462A1 (en) | 10 kv static load model parameter identification method based on similar daily load curves | |
| CN109559310B (en) | Power transmission and transformation inspection image quality evaluation method and system based on significance detection | |
| CN110222787A (en) | Multiscale target detection method, device, computer equipment and storage medium | |
| CN110070530A (en) | A kind of powerline ice-covering detection method based on deep neural network | |
| CN107564025A (en) | A kind of power equipment infrared image semantic segmentation method based on deep neural network | |
| CN107133960A (en) | Image crack dividing method based on depth convolutional neural networks | |
| CN108648169A (en) | The method and device of high voltage power transmission tower defects of insulator automatic identification | |
| CN110490252A (en) | A kind of occupancy detection method and system based on deep learning | |
| CN104992223A (en) | Dense population estimation method based on deep learning | |
| WO2015055168A1 (en) | Method and device for the automated early detection of forest fires by means of the optical detection of clouds of smoke | |
| CN109816625A (en) | A kind of video quality score implementation method | |
| CN109214308A (en) | A kind of traffic abnormity image identification method based on focal loss function | |
| US20210109140A1 (en) | Method for identifying parameters of 10 kv static load model based on similar daily load curves | |
| CN107392880A (en) | A kind of imitative pattern painting automatic generation method | |
| CN106997475A (en) | A kind of insect image-recognizing method based on parallel-convolution neutral net | |
| CN109191428A (en) | Full-reference image quality evaluating method based on masking textural characteristics | |
| CN110493803B (en) | Cell scene division method based on machine learning | |
| CN112767385B (en) | No-reference image quality evaluation method based on significance strategy and feature fusion | |
| CN108693441A (en) | A kind of electric transmission line isolator recognition methods and system | |
| CN107146216A (en) | A kind of non-reference picture method for evaluating objective quality based on gradient self-similarity | |
| CN109117723A (en) | Blind way detection method based on color mode analysis and semantic segmentation | |
| CN108876144B (en) | Transformer substation pre-addressing method based on deep learning algorithm |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |